When recording a color video image onto a digital magnetic tape, a conventional digital video cassette recorder (VCR) uses one of several compressive encoding methods to promote and improve recording efficiency. One standard compressive method is discrete cosine transform (DCT) variable-length encoding in which the number of output bits varies according to picture complexity. U.S. Pat. Nos. 4,831,440 and 4,939,583 respectively illustrate examples of the DCT and variable-length encoding techniques. However, when recording digital data onto a digital magnetic tape, fixing the number of output bits to a predetermined number of bits is very important.
It will be appreciated that the number of bits is a critical factor during variable-speed playback operations such as a high speed search operation. Since a playback head scans a plurality of tracks by traversing them during the high speed search operation, the output gain of the playback head decreases when it passes over the area of the tape between the recorded tracks. Thus, the playback head produces a diamond-shaped waveform envelope. Therefore, in the conventional recording and reproducing method, in order to arrange the minimum required video data necessary to reproduce a recognizable video image from a region scanned by a playback head, i.e., from a reproducible region video data is mixed so that it can be rearranged in a track or data for several tracks is rearranged into different tracks during a variable-speed playback operation, e.g., during a high speed search. However, in the conventional method, since the regions of the digital magnetic tape scanned by the playback head are different for each track, the format for rearranging the video data must be different for each track. This necessitates the use of a complicated rearrangement algorithm and, consequently, complicated circuitry for implementing the rearrangement algorithm.